Differences in weight status among Australian children and adolescents from priority populations: a longitudinal study

Childhood and adolescent overweight and obesity remain a global public health concern, with evidence of plateauing rates in many high-income countries, including Australia. Despite this, substantial inequalities persist: higher prevalence is observed among culturally and linguistically diverse (CALD) populations, Aboriginal and Torres Strait Islander peoples, and socioeconomically disadvantaged families. Nearly half of Australians were born overseas or have parents born overseas, and ongoing migration increases the nation’s cultural and linguistic diversity. Complex interrelationships among cultural, socioeconomic, environmental, and behavioral factors can influence children’s adiposity differently across developmental stages. Prior evidence suggests a “double disadvantage” of ethnicity and socioeconomic position in relation to overweight and obesity, and some priority populations experienced rising obesity rates between 1997 and 2015. However, gaps remain regarding which cultural and ethnic groups are at higher risk across childhood and adolescence, and at what developmental stages disparities emerge. The purpose of this study was to identify differences in BMI z-scores (zBMI) by cultural/ethnic groups and socioeconomic position across early childhood (2–5 years), middle childhood (6–11 years), and adolescence (12–19 years) using Australian longitudinal cohort data from 2006–2018.

Previous cross-sectional and longitudinal studies in Australia have reported higher odds or prevalence of overweight and obesity for certain priority populations, including immigrants from low- and middle-income countries and Aboriginal and Torres Strait Islander children, compared to peers from English-speaking or high-income backgrounds. The concept of a “double disadvantage” indicates compounded risk from both ethnicity and socioeconomic position. Evidence from school surveys and national datasets corroborates this pattern, though consistency varies by group; for example, findings for children from European backgrounds are mixed. Factors posited to underlie disparities include cultural and social norms regarding weight and activity, lifestyle behaviors, genetic variability, acculturation processes, and transitions from traditional to Western diets. Despite these insights, no prior studies examined longitudinal differences in zBMI from age 2 through 19 across specific cultural/ethnic groups and socioeconomic strata to pinpoint when disparities arise during child development.

Study design and data source: Longitudinal cohort analysis using the Longitudinal Study of Australian Children (LSAC) “baby” (B) and “kindergarten” (K) cohorts. LSAC is a nationally representative study employing a two-stage clustered sampling design from the Medicare enrolment database, with biennial data collection beginning in 2004. Sample: Waves 2–8 for both cohorts were used, yielding 9,417 children aged 2–19 years. Analyses were stratified by developmental periods: early childhood (2–5 years; B cohort waves 2–3), middle childhood (6–11 years; B waves 4–6 and K waves 2–4 pooled), and adolescence (12–19 years; B waves 7–8 and K waves 5–8 pooled). Outcome: BMI was calculated from measured height and weight by trained interviewers at each wave. Height used Invicta stadiometer (waves 2–3) and laser stadiometer (waves 4–8); weight used HoMedics digital scales (waves 2–3) and Tanita body fat scales (waves 4–8). BMI was converted to age- and sex-adjusted zBMI using WHO growth standards (≤5 years) and WHO reference (5–19 years). WHO SAS macro was used; biologically implausible zBMI values (<−5 or >5) were excluded. Exposures: Priority populations were defined as: (1) cultural and ethnic groups and (2) socioeconomic position (SEP). Cultural/ethnic groups followed Australian Bureau of Statistics classifications and included nine groups: English-speaking countries; Middle East and North Africa; East and South-East Asia; South and Central Asia; Europe; Sub-Saharan Africa; Americas; Oceania (excluding Australia and New Zealand); and Aboriginal and Torres Strait Islander peoples. Classification used child/parent country of birth, main language spoken at home (child in K cohort, primary and secondary parent), and Aboriginal and Torres Strait Islander status, with decision rules informed by ABS standards and migration patterns. SEP combined parents’ education, occupation, and family income into a z-score categorized into quintiles. Statistical analysis: Multilevel mixed linear regression models with repeated measures (level 1: time-specific zBMI; level 2: individual) estimated associations between exposures and zBMI within each developmental period. Cultural/ethnic group analyses were unadjusted; SEP analyses were adjusted for cultural/ethnic group as a confounder. Regression coefficients (β), 95% confidence intervals (CIs), and ICCs were reported. Survey weights were not applied due to pooling cohorts; sex-stratified analyses were not performed due to small subgroup sizes. Imputation was not conducted given low missingness for zBMI and SEP.

- Sample and follow-up: 9,417 children contributed 50,870 zBMI measurements and approximately 101,740 person-years of follow-up. About 75% were from English-speaking households. - Cultural/ethnic group associations with zBMI (referent: English-speaking): • Early childhood (2–5 years): Higher mean zBMI in Middle East & North Africa (β = 0.33; 95% CI: 0.16, 0.49) and Americas (β = 0.30; 95% CI: 0.01, 0.59). Lower zBMI in South & Central Asia (β = −0.58; 95% CI: −0.78, −0.37). • Middle childhood (6–11 years): Higher zBMI in Oceania excluding Aus/NZ (β = 0.59; 95% CI: 0.40, 0.78), Middle East & North Africa (β = 0.42; 95% CI: 0.28, 0.56), Americas (β = 0.41; 95% CI: 0.18, 0.63), and Aboriginal & Torres Strait Islander (β = 0.13; 95% CI: 0.01, 0.26). Lower zBMI in South & Central Asia (β = −0.28; 95% CI: −0.43, −0.12), East Asia (β = −0.11; 95% CI: −0.21, −0.02), and Africa (β = −0.23; 95% CI: −0.47, 0.01; not statistically significant at 0.05). • Adolescence (12–19 years): Patterns broadly similar to middle childhood, with notably higher effect for Aboriginal & Torres Strait Islander children (β = 0.30; 95% CI: 0.14, 0.46). - Socioeconomic position (adjusted for cultural/ethnic group): Clear socioeconomic gradient in zBMI with greater disadvantage associated with higher zBMI. Compared with the most advantaged quintile (Q5): • Q1 (most disadvantaged): β = 0.10 (95% CI: 0.02, 0.18) in early childhood; β = 0.22 (95% CI: 0.17, 0.27) in middle childhood; β = 0.23 (95% CI: 0.18, 0.29) in adolescence. • Q4: β = 0.03 (95% CI: −0.04, 0.10) in early childhood; β = 0.04 (95% CI: 0.01, 0.08) in middle childhood; β = 0.05 (95% CI: 0.01, 0.09) in adolescence. - Overall, children from Middle East & North Africa, the Americas, and Oceania had consistently higher zBMI than the English-speaking referent across childhood stages, while children from South & Central Asian and East Asian groups had consistently lower zBMI; European children were similar to the referent.

Findings demonstrate culturally patterned disparities in weight status across childhood and adolescence in Australia. Children from South & Central Asian, East Asian, and African households generally had lower zBMI than English-speaking peers, whereas those from Middle East & North Africa, Oceania (excluding Australia and New Zealand), and the Americas had higher zBMI across all developmental stages. Aboriginal and Torres Strait Islander children exhibited lower zBMI in early childhood but higher zBMI in middle childhood and adolescence. A consistent socioeconomic gradient was evident, with greater disadvantage associated with higher zBMI at all ages. These results highlight the need for targeted, culturally relevant prevention strategies that consider both cultural/ethnic background and socioeconomic context, and the optimal timing of interventions. For example, early childhood prevention may be less suitable for Aboriginal and Torres Strait Islander children (who have among the lowest zBMI at that age), while preconception, pregnancy, and early infancy programs may benefit groups with higher zBMI at ages 2–5. Middle and high school programs may be optimal for Aboriginal and Torres Strait Islander, Americas, and Middle East & North African households. Policymakers could adapt existing programs or implement strengths-based, community-led approaches to address disparities. The study’s results align with prior cross-sectional and longitudinal research in Australia showing higher overweight/obesity among Middle Eastern/North African, Oceanian, and Aboriginal and Torres Strait Islander children and lower prevalence among Asian children, while evidence for European-background children remains mixed.

The prevalence of overweight and obesity among Australian children is culturally patterned, with clear disparities across priority populations and a strong socioeconomic gradient. Identifying which groups are at higher risk at specific developmental stages can inform allocation of resources and the design and timing of culturally tailored healthy weight programs. Culturally adapted interventions across pre-primary, primary school, and adolescence are recommended to reduce disparities in overweight and obesity among priority populations.

- Lack of self-reported ethnicity or ancestry in LSAC; cultural/ethnic classification relied on country of birth and language spoken at home, potentially causing misclassification (though 98% were classified using two simple decision rules). - Small sample sizes in some groups (e.g., Africa, Americas) limited further subgroup disaggregation and precluded sex-stratified analyses. - Survey weights were not used due to pooling of B and K cohorts; findings may not be fully representative of the Australian child and adolescent population.